A Corpus-Driven Analysis of Spoken and Written Academic Collocations

Abstract

As already well-acknowledged in the field, formulaic sequences often make up an important part of lexical knowledge. Within the L2 learning contexts, there is increasing demand for the need to understand and retrieve lexical items as in prefabricated chunks. In tertiary contexts, knowledge of academic collocations is even more required. For individual word forms, there is the General Service List (West, 1953), and the Academic Word List (Coxhead, 2000). However, what the field currently lacks is a list of academic collocations. Identification of collocations was conducted with the British Academic Spoken English and the Academic Corpus for comparison of academic collocations in spoken and written discourse. There was use of 20 node words in each corpus, which produced 934 written and 460 spoken collocations. With more number of written collocations being produced from the corpus, the study provides a comparative view of the top 50 collocations which suggests a concentration of common collocations in the field of economics. Within the guided approach for data-driven learning, there is proposal for the need to facilitate psychological conditions for learning collocations.